@atoms-tech/atoms-mcp 0.74.0 → 0.89.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/README.md CHANGED
@@ -31,7 +31,7 @@ ATOMS MCP gives AI agents direct access to your requirements, test cases, and tr
31
31
  - Generate traceability diagrams as Mermaid graphs
32
32
  - View project compliance dashboards
33
33
 
34
- Every action is logged with AI actor attribution (`mcp_claude`) and enforces your organization's role-based permissions.
34
+ Every action is logged with AI actor attribution (`mcp_claude`, `mcp_cursor` or `mcp_copilot`, from the client you connect with) and enforces your organization's role-based permissions.
35
35
 
36
36
  ## Available Tools
37
37
 
@@ -262,14 +262,14 @@ Override the default 60 requests/minute per-user rate limit. Set to a positive i
262
262
 
263
263
  ### `ATOMS_CLIENT_NAME`
264
264
 
265
- Identifier persisted to the audit log so you can see which client (e.g., `cursor`, `claude-desktop`, `ci-coverage-bot`) made each call.
265
+ Identifier persisted to the audit log so you can see which client (e.g., `cursor`, `claude-desktop`, `ci-coverage-bot`) made each call. When it is unset, the name your MCP client sends when it connects is used, else `claude_desktop`. Names are stored as at most 64 characters of `a-z`, `0-9` and `_`.
266
266
 
267
267
  ## Security
268
268
 
269
269
  - **Org-isolated policy** — Every config knob set by Org A's admin lives in a row scoped to Org A's `org_id` and is invisible to Org B. RLS on the `mcp_policy` table is the boundary; the admin UI is purely a hide.
270
270
  - **Row Level Security (RLS)** — All queries run with the user's JWT. The server never uses a service role key. You can only access projects your organization has granted you access to.
271
271
  - **Role enforcement** — Write tools check your org role (viewer, editor, admin) before making changes. Viewers are blocked from mutations.
272
- - **Audit logging** — Every tool call is logged with tool name, parameters, duration, and session ID. Write operations log to `change_history` with `actor: "mcp_claude"`.
272
+ - **Audit logging** — Every tool call is logged with tool name, parameters, duration, and session ID. Write operations log to `change_history` with the client's actor (`mcp_claude`, `mcp_cursor` or `mcp_copilot`).
273
273
  - **Rate limiting** — Built-in per-user sliding-window rate limiter (default 60 req/min, configurable).
274
274
  - **Tool annotations** — Every tool declares `readOnlyHint`, `destructiveHint`, `idempotentHint`, `openWorldHint` per the MCP spec, so capable hosts can surface accurate consent prompts.
275
275
  - **Input caps** — Every string field has an explicit upper bound (body 50k, summary 2k, names 64, etc.) to prevent embedding-cost and memory-exhaustion attacks.